Amp with high pitched static noise

[Diabolical Artificer]

That was a long way back John, but yes, it is dire, but I posted it to give the OP a rough idea, the pics so far indicate a similar simple amp with a low component count.

Andy.

[spacebiscuit]

Originally Posted by Diabolical Artificer

With the amp off & no fuse in the fuse holder, meter set to resistance, one lead to ground, the other lead on one side of the fuse holder, what you get? If the answer is under 10r you have a short.

I think I've done this right - I studied also watched a few videos showing how
to test for a short.

With amp powered down and p-mosfets removed i:

- touched my black probe when the earth is connected
- touched red on either side of both fuses as instructed.

With DMM set to 2m resistance the results were 0.100, 0.550, 0.500 and .050 - see attached photo. If I try lower resistance settings the display just shows '1'.

It looks as if the negative rail supplies the p-mosfets and the positive side the n-mosfets if I am not mistaken?

By the way, MOSFET were definitely the right way around, exactly how they were removed. The two boards are a copy of one another rather than mirrored - such that on could be swapped out for the other.

Do my results indicate a high resistance and therefore no short?

[Diabolical Artificer]

With DMM set to 2m resistance the results were 0.100, 0.550, 0.500 and .050 - see attached photo. If I try lower resistance settings the display just shows '1'

As there are no qualifiers or units given that I recognise it's hard to say what those figures mean.

At a guess "2m" is 2M = 2 mega ohms scale, so 0.100 could = 100k but then "2m" may mean 2 milli ohms. To save guess work can you measure a resistor, if you have one with a a pic of the meter please?

Andy.

[spacebiscuit]

I've measured the same resistor on 3 different settings.

[Diabolical Artificer]

Ok, that makes more sense, notice the capital M 0n "2M" indicating mega ohms. We are interested in any resistance lower than 200 or 200 ohms when looking for shorts, non of the readings as far as I can tell show a short between the power rails & ground.

I feel we need to step back a bit on this repair as I'm having trouble figuring out which fet pin goes where on which bit of PCB: it's hard trying to correlate all the info & pics on a thread this length. It's also very hard trying to remote fix an amp from 100's of miles away.

So, to try & sum up, you have four good mosfets, as far as we know but with the mosfets fitted a fuse blows. Take the mosfets off, then power it up, what happens? Before that though, I've just been back through the thread looking at pics of the board, where is the fuse or fuses? Can you take a pic of the power supply please?

Andy.

[G6ONEDave]

Could you also add a picture as in your post 82 but without the overlaid graphics please. I was looking at the tracks but they are obscured in some important areas.

Andy the fuse holders are top centre in the pic in post 82, just right of the volume control. The left hand one has that ceramic cap shorting out a couple of tracks.

I am still suspicious of that ceramic cap but we will see. An easy way to confirm if the cap is connected correctly is to check the positive volts fuse holder connections. Both fuse holders should have the same circuitry connected apart from the fact that one is positive and the other is negative.

I now suspect that the short occurs when the mosfets are powered in circuit. As in the mosfets are acting as an on/off switch and shunting the supply to earth rather than waiting for for an audio signal to be controlled by.

Don't refit the mosfets for now as there is still some further testing to do, such as comparison voltages between the working and non working channels.

Dave

[G6ONEDave]

Please also post a photo of the amp pcb from the component side, this will help in component positions and any wire links. Perhaps also do the component side but image reversed left to right. This will help with what is where when compared with the track view. These photos will be a big help in distance diagnosis, as we don't have the actual pcb in front of us. Thanks.

Dave

[Diabolical Artificer]

Well spotted Dave, it says "fuse2 in big letters. Here is the pic from way back of the top side of the PCB, see attached, I can't see a fuse holder so wondered if the fuse was not on this PCB if the fault lies elsewhere.

Andy.

[Richardgr]

Weren't there 2 PCBs that were identical, one working?

[Diabolical Artificer]

Weren't there 2 PCBs that were identical, one working?

Yes.

Right, fuse holders are the black tubes, something else someone pointed out whilst chatting yesterday is that there are four 1/4" jacks, could this mean there are two OP channels on one board bridged? And that the fets as connected at present are wrong? Should it be one P fet & N fet on each side of the board instead of two P's on the left, two N's on the right?

We're back for the need of a schematic again, I've not tried to analyse the board in great detail as I've been trying to deal with things as they occur, putting out the fires as it were but not finding out how the fire started.

Andy.

[spacebiscuit]

So the current status is:

4 x good mosfets on the working channel

4 x mosfet of unknown status on the channel in question. I'm pretty sure 2 of the replacment p-mosfets are either blown or of inferior quality, possibly one of the original n-mosfet has also blown but need to test.

When I power up without the P-mosets fitted the fuses do not blow. When added the P-Mosfet fuse blows.

I have a set of 4 mosfets (n and p) on order which the seller assures me are original.

Images attached as requested - the P-mosets go where the board is marked TR3 and TR4 - see 2nd image.

[G6ONEDave]

Thanks for the extra large pics. Looking at the tracks and component layout, this is one pcb of a pair. There are 4 mosfets per channel. In the first pic of post 91, the Nfets should be on the right and the Pfets should be on the left. It is helpful that the tracks can be seen through the pcb as it were.

The two sockets on the left in the first pic of post 91 are the audio input and the two sockets on the right are the audio output.

I'm happier re that ceramic cap, as there is one on the other fuseholder. I'd not noticed the second one before. The tracking around each fuseholder is slightly different to each other, also adding to a little confusion.

My take on that small sub panel is that it produces a little more gain to the circuit. Probably an oem modification.

I'll have a ponder and hopefully get back with some more suggestions.

Dave

[G6ONEDave]

I was wrong about that small panel as on closer inspection it has a scr on it. This small pcb must be a protection circuit. Have you measured this panel for being short circuit? Think it's the brown and orange wires that sense a voltage across the left hand 0.22 ohm resistor in first pic of post 91. It's a guess but I suspect that in the event of a fault the scr shorts a supply to 0v and blows the fuse, thus preventing the amp from working.

Be interesting to see what Andy and others think.

Dave

[Analogue man]

I was just wondering about this SCR and my thoughts are that it would be worth taking that out of circuit and testing it for a short or very low resistance between two wires.

[spacebiscuit]

The small board has "cutoff issue 1" printed on it. My initial suspicion were that the fault must lay here but I switched this over to the working channel and it worked so concluded that is was fine. Also the component all give the same reading as the other channel's.

[Analogue man]

OK not that then.

[G6ONEDave]

Well if nothing else it eliminates that small pcb from being the cause of fuse blowing. So it's back to the drawing board then and more pondering.

Dave

[G6ONEDave]

What type of diode is D1 and is it OK? Can't tell if it's a signal diode like a 1N4148 or a zener diode. It appears to be connected to the fet gates.

Dave

[spacebiscuit]

Originally Posted by G6ONEDave

What type of diode is D1 and is it OK? Can't tell if it's a signal diode like a 1N4148 or a zener diode. It appears to be connected to the fet gates.

Dave

Hmm, tested the D1 on both boards, both return "diode" but results differ, could this be something?

First image from the good channel, second from the other.

[Diabolical Artificer]

The results look fine, maybe the first showing capacitance is indicative of a fault, maybe not. what does D1 do? Biasing? Protection?

What this amp needs is a systematic approach rather than a component level approach. Having a schematic would really help too as has been mentioned loads of times.

The standard approach for something like this fault is to find any potential shorts, eliminate them, then power the DUT (device under test) with a current limited supply to prevent any more damage. Along side that you need to learn how the amplifier works, what connects to what etc. Forgive me if I sound exasperated, I feel like were going round in circles & getting nowhere fast.

Andy.